% fortune -ae paul murphy

Trust but verify: evaluating the message, not the messenger

Sometime, I'm guessing in the last eighteen months or so, internet usage passed an important milestone. Nobody rang bells or blew trumpets when it happened, but somewhere during that period the value, volume, and diversity of news and analysis posted by interested amateurs increased to the point that it became practical to live by the ideal that one should always judge the message, never the messenger.

What brought this to my attention seemed innocent enough: I cited a report by somebody named Laura as as authoritative on opinion pieces written by WWF, Greenpeace employees, and other Malthusians being passed off as science by the UN's IPCC - and then realized that not only do I know nothing about this person, but that the work speaks sufficiently for itself that there's no reason to care.

A lot of people don't think about stuff like this, but ever since Descartes published his Discourse on Method in 1637 formalizing work by Bacon, Galileo, and others on the scientific method and the rejection of authority as a source of knowledge, we've known how things should be done, but not had the means to do it outside the narrow bounds of whatever scholarship we may ourselves have been able to practice.

Now we have - and of course that's the way it should always have been but knowing what to do and being able to do it have hitherto been different things. Now they're not: a case of science fiction made real by the uncoordinated efforts of tens of thousands of people.

Unfortunately there's a corollary: a moral imperative: because if we can, we must.

And, of course, like all good moral imperatives this one comes with a catch: because negative authority still discounts. The legal maxim "falsus in unum, falsus in omnia" says it all: if you catch your source in one lie, you're entitled to assume anything else he says is a lie.

Mistakes and lies can be hard to tell apart, but inconsistency tells the story. Thus when a well known tech writer publishes something mildly critical of IBM's mainframe business model and subsequently seems to become both obsessively anti-Sun and obsequiously pro-IBM, the inconsistency suggests a failure to recognize the moral imperative to stick to the right - admittedly a conclusion I jump to in part because of my own unhappy experience in getting over 2,700 pieces of personalized hate mail for debunking IBM's mainframe Linux as over priced, over hyped, lunacy.

The bottom line, though, is that this change has happened: that internet access now means that message generally trumps messenger in both theory and practice - meaning, among other things, that anybody who still acts as if attacking the messenger's credentials; trying, for example, to dismiss the content of the Wattsupwiththat climate site by claiming that Mr. Watts isn't a scientist, is categorizing himself as intellectually weak, emotionally dishonest, and fundamentally out of touch with reality.


Paul Murphy wrote and published The Unix Guide to Defenestration.